Same on ComfyUI, usually i only get this line:

"ERROR lora diffusion_model.input_blocks.4.1.transformer_blocks.0.attn2.to_k.weight shape '[640, 2048]' is invalid for input of size 491520" (and a whole bunch of similar errors after it), if i accidentally load an SD 1.5 model on SDXL/PonyXL.

Maybe the uploader accidentally put Pony instead of 1.5?

Edit: I had originally put here that maybe AutismMix was the issue, since most of the images weren't using it, and had asked if you were also using it. It's probably not, since i just saw an image using AutismMix

@Genie123 Strangely enough there's no errors when launching WebUI. The loras just isn't in the tab (have just discovered another three not loading) 😒

I have a similar problems, but there are things you can do to bypass it...

1: you can force-call it by putting the lora name in your prompt <lora:Bulge_Mania:1> yourself
or...
2: in your setting, search lora and tick the box next to Always show all networks on the Lora page, apply and refresh your lora tab and it should then show up

This may be a late reply, but I wanted to let you know that I was able to solve the same problem I was having.

Perhaps the LORA metadata is loaded as SD1 and therefore invisible?

Try loading the SD1 checkpoint, updating the LORA list and checking if it is visible.

If it does, change the “Stable Diffusion version” from SD1 to SDXL or Unknown and you should be able to see it next time.

I had this problem with another LORA and after solving it this way, I was able to use this LORA which was previously invisible.

Translated with DeepL.com (free version)

The model page is incorrectly listed as a Pony model.
It is in fact a SD v1.5 model.
The metadata for the file says "ss_base_model_version": "sd_v1"

Correct, the model page is incorrectly listed. It is in fact a *SD v1.5* model.
The metadata says "ss_base_model_version": "sd_v1",